Reducing Costs and Variations in Care

Danbury Hospital recently embarked on a cost efficiency project to record, report, and analyze individual physician performance to reduce unnecessary inpatient resource use and variations in care. Danbury’s endeavor, dubbed the DRG Cost Efficiency Project, ended up saving the institution $6 million.

Danbury Hospital is a 371-bed regional medical center and university teaching hospital associated with Yale University School of Medicine, the University of Connecticut School of Medicine, and the University of Vermont. Danbury drove its DRG Cost Efficiency Project with a mandate from the Joint Commission's Ongoing Professional Practice Evaluation (OPPE) standards to measure practitioner performance, which the hospital did by using physician management software from the Austin, Texas-based Crimson, a division of The Advisory Board Company. The management software feeds off of data from the hospital’s Soarian electronic health record (from the Malvern, Pa.-based Siemens), aggregates and reports it.

“You can talk a lot about [performance] in a static way, saying, ‘don’t do unnecessary things,’ but unless you have a good way in measuring what it is they’re [physicians] doing collectively or as individuals, and reporting it back to them, it’s hard to move the dial,” says Matthew Miller, M.D., chief medical officer at Danbury Hospital. Miller notes that a key function of this particular physician management tool was that it risk adjusted the data to allow informaticists to aptly compare similar types of patients and adjust data for diagnosis, severity, and complication rates.

For Danbury’s DRG project, eight high-volume diagnosis-related groups (DRGs) were targeted, within several broadly important clinical areas like congestive heart failure, chronic obstructive pulmonary disease (COPD), joint replacement, and stroke. The eight DRGs were identified within those areas because of their connections to high-volume, high-cost procedures. Indicators like average length of stay (ALOS) and readmission rates were also evaluated.

Miller notes that the DRG project was to change the practice of what he calls “defensive medicine,” where all possible tests are ordered to rule out certain diagnoses, but as Miller puts it “doing more things is actually not only not better, it actually might be worse.” He adds, “So if you do unnecessary tests you might get false positives and end up with unnecessary biopsies, and unnecessary treatments, not only you’re not helping the patient, you might hurt them.”

Identifying Targets, Streamlining Order Sets
Miller worked with his CMIO Eric Jimenez, M.D., Maribeth Cross, R.N., performance improvement coordinator, and chief physicians in the eight prescribed departments to identify unnecessary variations in care and establish new target rates for core measures. Each section head identified six to eight targets to focus on in each group and selected goals for the year to compare with the baseline. Miller and Deborah Geambazi, R.N., senior analyst, clinical data, then monitored progress and met individually with physicians on a monthly basis to go over individual metrics.

“Just [having] an awareness is critical,” Miller says. “And giving doctors back data that showed them how they did last quarter or last month, what progress you’re making or not, but also notice that if you [make] progress, you’re not seeing a deterioration in quality.”

The hospital also streamlined order sets to limit the amount of tests and orders for each diagnosis. For example Miller says that before the project, the stroke order sets for the first and second inpatient days had every conceivable drug and test a physician could possibly order. Miller and his team restricted some of the orders, necessitating approval for certain tests and drugs. Within 12 months each DRG developed mid course corrections if the original target was too aggressive. Also, a dashboard was reviewed quarterly by senior hospital leadership and Danbury’s board of directors, and outcomes for the DRG project became part of the operational goals for the hospital.

Successful Results, Project Expansion
In its first year, Danbury reduced the 30-day readmission rate for stroke patients by 10 percent and average length of stay by 8 percent, as well as trimmed actual average charges by 12 percent, among other key measures. The institution’s overall savings was $2.9 million, and when all related diagnoses were examined, like all patients with a respiratory or cardiology diagnosis, the financial savings was a whopping $6 million.

The success of the DRG project instigated a broader look at other opportunities to improve utilization in the hospital. Danbury is now looking horizontally at its costs, inspecting radiology and laboratory—specifically CAT scans and blood tests—to identify where resources can be conserved. Danbury will also investigate the pharmacy formulary, to see if less expensive drugs can be substituted when appropriate. More diagnoses are being added to the project, and complementary technology like supply chain products are being purchased to supplement the IT efforts.

Miller sees this project fitting into Danbury’s larger goals to set up a pay for performance incentive for its hospitalists and to create an accountable care organization to help collect data in one platform across the continuum of care. “A lot of information technology in healthcare is very good at what happened to that patient but is not good at trending,” says Miller. “It doesn’t say what we do with all the other diabetics.”